Professional Documents
Culture Documents
Clinical Psychological Science 2013 Yang 5 15
Clinical Psychological Science 2013 Yang 5 15
Published by:
http://www.sagepublications.com
On behalf of:
Additional services and information for Clinical Psychological Science can be found at: Email Alerts: http://cpx.sagepub.com/cgi/alerts Subscriptions: http://cpx.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> Version of Record - Dec 14, 2012 OnlineFirst Version of Record - Nov 15, 2012 What is This?
Empirical Article
Clinical Psychological Science 1(1) 515 The Author(s) 2013 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/2167702612464618 http://cpx.sagepub.com
Abstract Abnormal perceptual experiences are central to schizophrenia, but the nature of these anomalies remains undetermined. We investigated contextual processing abnormalities across a comprehensive set of visual tasks. For perception of luminance, size, contrast, orientation, and motion, we quantified the degree to which the surrounding visual context altered a center stimuluss appearance. Healthy participants showed robust contextual effects across all tasks, as evidenced by pronounced misperceptions of center stimuli. Schizophrenia patients exhibited intact contextual modulations of luminance and size but showed weakened contextual modulations of contrast, performing more accurately than controls. Strong motion and orientation context effects correlated with worse symptoms and social functioning. Importantly, the overall strength of contextual modulation across tasks did not differ between controls and schizophrenia patients. In addition, performance measures across contextual tasks were uncorrelated, implying discrete underlying processes. These findings reveal that abnormal contextual modulation in schizophrenia is selective, arguing against the proposed unitary contextual processing dysfunction. Keywords schizophrenia, center-surround, contextual effects, perception decit, visual processing
Received 4/24/12; Revision accepted 9/21/12
Impairments in visual processing are pervasive in schizophrenia, ranging from deficits in simple visual discriminations (Butler & Javitt, 2005) to abnormalities in high-level visual perception (Kim, Park, & Blake, 2011). These visual deficits may exacerbate and/or lead to cognitive and social problems (Uhlhaas & Mishara, 2007). Despite many reports of visual deficits in schizophrenia, their role in the symptoms and etiology of the disorder remains unclear. One obstacle to understanding the nature of visual abnormalities in schizophrenia is the concurrent presence of other performance impairments. For example, well-established attentional and working memory deficits (Lee & Park, 2005) may cause poor performance in a variety of visual tasks, making it difficult to isolate specific visual deficits. Remarkably, abnormalities in visual processing in schizophrenia do not always show up as deficiencies in task performance. Rather, visual perception in schizophrenia appears to be qualitatively different, characterized by performance deficits or performance enhancements (Barch et al., 2012; Dakin, Carlin, & Hemsley, 2005; Tadin et al., 2006; Uhlhaas, Phillips, Mitchell, & Silverstein, 2006; Yoon et al., 2009; Yoon et al., 2010). The findings of enhanced performance in schizophrenia
are difficult to attribute to nonvisual factors and thus provide a direct behavioral measure of underlying visual abnormalities. Importantly, better-than-normal visual performance in schizophrenia seems to be restricted to tasks of spatial contextual effects in visual perception, wherein the perceptual appearance of a visual feature depends on its surrounding spatial context. For example, in the well-known Ebbinghaus illusion (Fig. 1e), a circle surrounded by larger circles appears smaller than the same circle presented alone. Similarly, the presence of a high-contrast background decreases the apparent contrast of smaller foreground features (Fig. 1b). Although these illusions provide striking examples of nonveridical visual perception, they are considered functionally advantageous because they serve to enhance differences among visual features and consequently facilitate their segmentation from their background (Albright & Stoner, 2002). These contextual illusions, in other words, arise from visions adaptive propensity to emphasize
Corresponding Author: Sohee Park, Department of Psychology,Vanderbilt University, 111, 21st Avenue South, Nashville, TN 37240-7817 E-mail: Sohee.Park@vanderbilt.edu
Yang et al.
Fig. 1. Illustrations of stimuli in each experiment. (a) Brightness induction illusion: the reference stimulus (left) with darker surround is perceived as being brighter than the target stimulus (right) with lighter surround. (b) Surround contrast illusion: A high-contrast pattern in the surround reduces the perceived contrast of the center stimulus (left) relative to a reference stimulus of equal contrast (right). (c) Surround motion and (d) orientation repulsion illusions: The orientation or motion direction of a center stimulus appears to be repelled away from the orientation or motion direction of the surrounding pattern (arrows denote motion direction). (e) Ebbinghaus size illusion: When a circle is presented with larger circles in the surround (right), the center circle is typically perceived as being smaller than a stimulus of equal size (left). Note: Scale bar denotes the stimulus display size in degrees of visual angle. The spacing between stimuli in 1a is not on the same scale as the size of the stimuli.
relative differences among features rather than their absolute characteristics. However, a neural implementation of such contextual interactions must strike a balance between enhancement of differences among features and veridical integration of visual signals (e.g., Braddick, 1993). Consequently, both abnormally weak and abnormally strong contextual modulations constitute departures from typical vision and are arguably maladaptive. As mentioned, recent studies suggest that individuals with schizophrenia (SZ) may be less influenced by visual context on some tasks, thus enabling them to perceive absolute characteristics of individual features more accurately. For example, Dakin et al. (2005) reported that SZ were better than healthy participants at judging stimulus contrast under conditions of contextual modulation. This tendency toward veridical performance by SZ, however, suggests processing impairmentsnamely, a weakening of inhibitory mechanisms that normally modulate perceived stimulus contrast (Butler, Silverstein, & Dakin, 2008). Yoon et al. (2009) extended these results, reporting abnormally weak orientation-specific surround inhibition, as evidenced by more accurate contrast discrimination in SZ. Yoon et al. (2010) subsequently linked this deficit with abnormally low GABA (gamma-aminobutyric acid) levels in the visual cortex, suggesting a possible causal role of impaired inhibitory mechanisms.
However, a recent study with a large sample size (Barch et al., 2012) failed to replicate the original results by Dakin et al. after controlling for differences in baseline performance. This result reveals a possible attentional confound and underscores that the nature of contextual contrast modulation in SZ remains an open question. In a similar vein, our group found abnormally weak surround inhibition in SZ tested on a motion perception task (Tadin et al., 2006), a deficit that involves improved detectability of large moving stimuli. In addition, SZ are less prone to the Ebbinghaus size illusion, exhibiting more accurate size perception when surrounding context is present (Uhlhaas et al., 2006). In summary, visual perception in schizophrenia is characterized by diminished contextual modulations (see Chen, Norton, & Ongur, 2008, for an exception), which paradoxically produces more accurate perception of absolute stimulus characteristics. Considered together, results from these studies suggest a generalized contextual processing deficit in schizophrenia, in line with contextual disturbances in schizophrenia reported in other sensory and cognitive domains (review by Phillips & Silverstein, 2003). These contextual processing deficits may be indicative of a unitary mechanism that is a core feature of schizophrenia (Cohen & Servan-Schreiber, 1992). The empirical evidence for this appealing hypothesis, however, is
Visual Context Processing in Schizophrenia sparse, given that prior studies have typically focused on single contextual tasks, several of which have yet to be replicated. Furthermore, there is considerable variability in the magnitude of observed contextual deficits within and across functional domains, and in some cases contextual abnormalities seem restricted to specific tasks (see Park, Lee, Folley, & Kim, 2003) and subsets of patients (Tadin et al., 2006; Uhlhaas et al., 2006). Nonetheless, the utilization of measures of contextual processing has already found its way into clinical testing (Gold et al., 2012). There has been a rapid growth in the use of context tasks in clinical trials and large-scale, NIH-supported studies of schizophrenia, including tasks focusing on visual context. Given this trajectory, it is imperative to specify and identify the conditions under which contextual processing is weakened in schizophreniaa goal requiring comprehensive investigations. We investigated a diverse set of psychophysical tasks in which spatial context is known to affect stimulus appearance, including both previously tested contextual modulations and modulations that have yet to be examined in SZ. Specifically, we investigated contextual effects in luminance, contrast, size, orientation, and motion perception. In all these tasks, weaker contextual modulation is associated with more accurate perception of visual features. Thus, if a broad contextual deficit exists, SZ should exhibit more veridical performance relative to healthy matched controls when the visual context is present. Moreover, if a common mechanism underlies contextual effects in these tasks, then the magnitude of impairment should correlate across tasks. In sum, we aimed to investigate the full range of visual context processing to elucidate the profile of impaired, intact, and enhanced functions in schizophrenia.
7 for the Assessment of Positive Symptoms (SAPS; Andreasen, 1984), and the Scale for the Assessment of Negative Symptoms (SANS; Andreasen, 1983). SZ were clinically stable at the time of testing, with a mean BPRS score of 13 (SD = 8), SAPS score of 14 (SD = 13), and a SANS score of 17 (SD = 7). There were no significant differences in age (p > .05) between SZ (M = 41, SD = 8) and CO (M = 39, SD = 9) or in the proportion of women (p > .05) between SZ (37%) and CO (48%). Although SZ (M = 13, SD = 2) had fewer years of education than CO (M = 16, SD = 2), t(51) = 4.4, p < .001, they did not differ in IQ (p > .05), with mean estimated IQs of 104 (SD = 9) for SZ and 106 for CO (SD = 11), as assessed by the National Adult Reading Test (Nelson, 1982). Finally, social functioning, as assessed with the Social Functioning Scale (Birchwood, Smith, Cochrane, Wetton, & Copestake, 1990) was worse in SZ (M = 111, SD = 9) than in CO (M = 123, SD = 5), t(43) = 5.4, p < .001. The Institutional Review Board of Vanderbilt University approved this study protocol. All participants provided written informed consent and were paid.
Apparatus
Stimuli were created with MATLAB and the Psychophysics Toolbox (Brainard, 1997; Pelli, 1997) and were presented on a linearized CRT monitor (1280 960 resolution at 120 Hz). Participants viewed the stimuli at a distance of 73 cm, with the head stabilized by a chin rest. The display background luminance was 35.2 cd/m2 except in the brightness induction task, where it was 0.11 cd/m2. The ambient illumination was 0.16 cd/m2.
Context battery
A battery of five psychophysical tasks was administered to assess contextual effects in a broad range of stimulus dimensions: luminance, contrast, size, orientation, and motion direction (Fig. 1). All tasks involved a center stimulus, of which the perceived appearance, as judged by healthy observers, was altered by the presence of surrounding stimuli. Participants were instructed to judge the appearance of the center stimulus by comparing it with a fixed reference stimulus. For each task, the point of subjective equality (PSE) was measured to quantify the magnitude of the contextual modulation. PSEs were estimated for the brightness induction task by method of adjustment and, for the remaining tasks, by adaptive staircases. With the exception of the motion task, stimuli were presented until a response was made. All tasks also included a no-context control condition, identical to the main context condition except that no surrounding context was present. These control conditions allowed us to establish that participants accurately judged stimulus dimensions tested in different tasks and to account for any baseline biases of participants.
8 Brightness induction task. Two circles (0.5 radius) surrounded by annuli (2.4 radius) were simultaneously presented 15 apart (Fig. 1a). The luminance of the reference circle displayed on the left was fixed at 6 cd/m2, whereas its surrounding annulus was 8, 12, 16, 20, or 24 cd/m2. Measuring brightness induction with several surround luminance values allowed us to also compare the pattern in surround modulation between groups. The luminance of the annulus on the right was fixed at 24 cd/m2, whereas the initial luminance of its inner circle (target) was randomly chosen from a range of 2 to 14 cd/m2. Participants then adjusted the luminance of the target circle on the right to match the luminance of the reference circle on the left. Luminance was adjusted with a pair of response keys that increased or decreased target luminance in step sizes of 0.2 cd/m2. To estimate PSEs, three such adjustments were performed for each surround luminance. Brightness induction strength was defined as the difference (in cd/m2) between fixed luminance of the reference circle (6 cd/m2) and the perceived (i.e., adjusted) luminance of the target (typically much higher). Surround contrast illusion task. The stimulus display (Fig. 1b) was similar to that used previously (Barch et al., 2012; Chubb, Sperling, & Solomon, 1989; Dakin et al., 2005). Stimuli consisted of two circular patches (1.67 radius; 13.5 horizontal center-center separation) of spatial frequency filtered noise (1 cycle per degree, center frequency; 0.25-octave bandwidth). The Michelson contrast of the reference patch was fixed (20%) and surrounded by a high-contrast noise annulus (6.67 radius, 97% contrast). The other patch was the target stimulus, the starting contrast of which was randomly chosen (10%30%). The positions (left or right) of reference and target stimuli were randomly assigned on each trial. Participants pressed one of two keys to indicate which patch appeared higher in contrast, and their responses were used to adaptively adjust the contrast of the target to match the apparent contrast of the reference stimulus. Ebbinghaus size illusion task. This task was a variant of the Ebbinghaus illusion (Fig. 1e). The fixed reference display consisted of a small circle (1.08 radius) surrounded by five evenly spaced large circles (2.17 radius, 0.75 apart). The target stimulus was a small circle, the initial size of which was randomly chosen (0.921.08 radius) and thereafter varied by the staircase procedure. The target and reference stimuli were presented 15 apart (center-center), and their positions (left or right) were randomly assigned on each trial. All stimuli were presented at 97% Michelson contrast. Participants pressed a key to indicate which of the two center circles was larger. Surround orientation repulsion task. The stimulus display consisted of a small circular grating (0.5 radius, 50% contrast, 3 cycles per degree) surrounded by a large, high-contrast annulus (4 radius, 97% contrast, 3 cycles per degree), both of random phase (Fig. 1d). The annulus orientation was fixed at 15 counterclockwise from vertical, whereas the center
Yang et al. orientation was either 11 clockwise or counterclockwise at the start of the task and thereafter determined by the staircase procedure. Participants used key presses to indicate whether the center patch appeared tilted clockwise or counterclockwise relative to vertical. Surround motion repulsion task. The stimulus display consisted of a stimulus moving within a small circular aperture (1 radius) surrounded by another stimulus moving within a large annulus (6 radius; Fig. 1c). Both stimuli moved at 3 deg/s and consisted of spatial frequency filtered noise (80% contrast; 1 cycle per degree, center frequency; 0.25-octave bandwidth). The surround motion direction was either 45 clockwise or counterclockwise from vertical, whereas the center motion direction was either 18 clockwise or counterclockwise at the start of the task and thereafter determined by the staircase procedure. To avoid pursuit eye movements, stimuli were presented for 200 ms and immediately replaced with a blank screen. Participants indicated whether the central motion direction was clockwise or counterclockwise relative to vertical.
Procedure
Each experiment consisted of four blocks, starting with the control block (without context) and followed by three context blocks. Each block consisted of two interleaved one-up/onedown staircases. The step size decreased after every two reversals. Each staircase converged after seven reversals, and the average of the last four reversals was taken as the PSE estimate; that is, the resultant PSE for each participant was an average of six staircases (two in control tasks). One exception was the brightness induction task, for which an adjustment method was used (described earlier). Because all the tasks were subjective judgments about stimulus appearances, feedback was not provided. There was no time limit in making a response. The order of experiments was randomized for each participant. The entire battery took about 1 to 1.5 hours to complete. Before each task, participants were given detailed instructions to make sure that they understood the task, and they performed five practice trials. The magnitude of contextual effects was measured by the PSE estimates with surrounding context relative to the performance on the control condition (i.e., as the degree to which a participants perception changed as a result of adding the surrounding context). The units for luminance, contrast, and size tasks were cd/m2, log10 contrast, and arcmin, respectively. Orientation and motion repulsion were measured as angular repulsions in degrees.
Psychometric properties
To ensure that we can appropriately estimate contextual processing strength, we considered a number of psychometric issues: ceiling effects, floor effects, and measurement reliability. None of the results approached stimulus-constrained
Visual Context Processing in Schizophrenia ceilings that are inherent in our tasks (e.g., 90 repulsion in the orientation task). In fact, all results were considerably weaker than the stimulus-constrained ceilings. Theoretical floors also exist and would be manifested as a no contextual effect for each task. However, as detailed in the Results section, all CO data sets exhibit strong contextual effects, leaving ample dynamic ranges to reveal weaker contextual effects in SZ. Finally, data for each task did not deviate from normality and equality of variance, as assessed with Kolmogorov-Smirnov test and Levenes test, respectively. To assess measurement reliability, we split each set of data into halves or thirds and correlated these partial data sets. For size and contrast tasks, we obtained six independent PSE estimates; thus, we split the data into halves. The CO/SZ split-half reliabilities were .90/.90 and .84/.83 for size and contrast experiments, respectively. For motion and orientation tasks, we obtained three pairs of measurements (with a pair consisting of two center directions/orientations). To estimate measurement reliability, we correlated the second and third estimates. CO/SZ correlations were .69/.79 and .58/.82 for motion and orientation tasks, respectively. The modest correlation for CO in the orientation task is largely due to one CO subject who did not show a contextual effect in one measurement; without him or her, the correlation is .71. Note that somewhat lower numbers in motion and orientation tasks are expected, given that only two thirds of
9 the data are used. Finally, CO/SZ correlations between second and third adjustment estimates in the brightness task were .87/.88. In sum, we found reliabilities for SZ and CO to be comparable for each task and relatively high. For SZ, all split-data reliabilities were between .79 and .90 (average = .84).
Results
Owing to experimenter error, data for a few tasks were missing for two SZ: size for one patient and motion, contrast, and brightness induction for both patients. However, their data on the other tasks were retained. Moreover, if a participants data fell three standard deviations or more from the group mean, his or her data were excluded. This resulted in the exclusion of eight data sets (one, three, and four data sets in the contrast, orientation, and motion tasks, respectively), accounting for approximately 3% of all data. Surround contextual effects were observed across all tasks and for both groups (one-sample t tests, all p .001). Moreover, CO and SZ did not significantly differ in baseline conditions in which surround stimuli were absent (independent samples Welchs t tests, all p > .05). This result shows that SZ had no problems accurately performing visual tasks in this study. In what follows, we report the results of each task (Fig. 2), followed by combined analyses across tasks (Fig. 3).
a
Change in Size (Arcmin)
b
Motion Direction Change
12 8 4 0
Motion Repulsion
c
4
Orientation Repulsion
3 2 1 0 CO SZ
8 4 0 CO SZ
CO
SZ
d
Contrast Change (log)
0.3 0.2 0.1 0
Contrast Illusion
*
e
Brightness Change (cd/m2)
10 8 6 4 2 0 5 10
Brightness Induction
}
CO SZ 15 20 25
Orientation Change
10 8 6 4 2 0 CO SZ
CO
SZ
Surround Luminance
Fig. 2.Results from five context modulation experiments: (a) Ebbinghaus size illusion, (b) motion repulsion, (c) orientation repulsion, (d) surround contrast illusion, and (e) brightness induction. In 2e, data for five surround luminance levels are shown in the left panel. The right panel isolates the data for the condition with the strongest surround modulation (8 cd/m2 ), a condition included in Figure 3.
10
Orientation Motion Size Contrast Brightness Contextual Modulation Index 0.4 0.2 0 0.2 0.4
Yang et al.
0.8 0.6
SZ z Score Relative to CO
Fig. 3. The magnitude of contextual effects in SZ. For each task, the magnitude of contextual effect in schizophrenic patients (SZ) was converted into z scores relative to the mean and variance of the control group (CO). The contextual modulation index (CMI) was computed for each participant as the average z score across tasks. Negative values indicate weaker contextual illusions in SZ, whereas positive values indicate stronger contextual illusions in SZ relative to CO. The only significant difference was weaker contextual modulation of contrast in SZ, t (42) = 4.87, p = .03. CMI for SZ was not statistically different from zero (z = 0.048, p = .96). Error bars denote the standard error of the mean of the z scores in the schizophrenia group, and the shaded region denotes the standard error of the mean of the control group.
The contextual effects of size, t(50) = 0.17, p = .68, Cohens d = 0.11, nCO = 23, nSZ = 29; motion, t(45) = 0.14, p = .71, d = 0.11, nCO = 21, nSZ = 26; and orientation, t(47) = 2.37, p = .13, d = 0.43, nCO = 23, nSZ = 27, in SZ were not significantly different from those of CO, according to independent samples Welchs t tests. However, SZ did show a weaker contrast illusion, t(42) = 4.87, p = .03, d = 0.64, nCO = 23, nSZ = 27. In the brightness induction task, a repeated measures analysis of variance with surround luminance as the within-subjects factor and with group as the between-subjects factor revealed a main effect of surround luminance, F(4, 196) = 324.3, MSE = 611.35, p < .001, but no main effect of group, F(1, 49) = 1.24, MSE = 8.87, p = .27, d = 0.31, nCO = 23, nSZ = 28, nor an interaction between luminance and group, F(4, 196) = 1.7, MSE = 3.17, p = .16. To examine the pattern of results across all contextual tasks, we normalized results for each task relative to the performance of the CO to derive z scores (Fig. 3). For the brightness induction task, we used the PSE estimate in the surround luminance condition that would evoke the strongest illusion (surround luminance of 8 cd/m2). Using a mixed model analysis of variance with task and group as fixed factors, we found no significant main effect of group, F(1, 50) = 0.11, p = .74, with a marginal main effect of task, F(4, 49) = 2.46, p = .06. The combined analysis yielded a significant interaction between group and task, F(4, 49) = 2.53, p = .05, indicating that the relative strength of contextual effects in SZ depended on which task was used. Figure 3 also shows that tasks for which SZ
were not significantly different from CO, when considered in isolation, are different from one another. Namely, in SZ, contextual modulation of orientation and motion are stronger than contextual modulation for both luminance, t(24) = 2.7, p = .01, d = 0.75; t(25) = 2.13, p = .04, d = 0.55, respectively, and contrast, t(24) = 3.59, p = .001, d = 1.0; t(24) = 2.27, p = .03, d = 0.62, respectively. Although most of these analyses would not survive multiple comparisons correction, they indicate that the direction of findings in contextual processing studies of SZ will likely depend on which tasks are used. To estimate a general measure of contextual processing in SZ, we derived a contextual modulation index (CMI) for each patient by averaging z scores across tasks (relative to CO). If patients show general weakening of contextual processing, then CMI should be negative. A positive CMI would indicate a general strengthening of contextual processing. The result, however, is a z value of 0.048, with an associated p value of .96 (Fig. 3). In other words, CMI is statistically zero. Furthermore, variance did not differ between groups, F(1, 51) = 0.43, p = .52, ruling out the possibility that a lack of CMI differences is due to equal numbers of SZ with abnormally strong and abnormally weak CMIs. We also examined intertask correlations to test whether a weak contextual effect on one task would predict a weak contextual deficit on other tasksa finding that would be expected by the presence of a common underlying deficit. However, no such correlations were found in SZ (all |Pearsons r| < .40, all
Visual Context Processing in Schizophrenia p > .06) nor in CO (all |r| < .41, all p > .07). Correlations did not reach significance even without Bonferroni correction. The strongest correlation in the SZ group performance was between orientation and motion tasks (r = .39, p = .07), whereas the rest of the intertask correlations fell below .18 (p > .38). This lack of intertask correlations further suggests that our contextual tasks reflect largely distinct neural mechanisms. Finally, we probed the relationships between the strength of contextual illusions and clinical symptoms in SZ. Results showed that the magnitudes of both orientation and motion illusions were positively correlated with the BPRS such that the severity of symptoms was associated with stronger illusory repulsion (Figs. 4a and 4b Spearmans rho [rs] = 0.50, p = .008, n = 27; rs = .57, p = .002, n = 26, respectively). Orientation and motion illusions were correlated with negative symptoms (SANS; rs = .46, p = .02, n = 27; rs = .41, p = .04, n = 26,
11 respectively) and positive symptoms (SAPS; rs = .38, p = .05, n = 27; rs = .44, p = .02, n = 26, respectively). Stronger orientation repulsion was further associated with longer illness duration (Fig. 4c; rs = .46, p = .02, n = 27) and lower social functioning (Fig. 4d; rs = .48, p = .01, n = 25). The same analyses were applied after excluding one to three additional outliers based on symptoms scores, but still all correlations remained significant. In fact, correlations reported with the BPRS and SANS grew stronger (rs = .50~.67, p .001~.01). BPRS, illness duration, and social functioning were not correlated (all |rs| < .25, all p > .20). Correlations with symptom scores and social functioning were not found with other tasks (all rs < .32, all p > .09). Furthermore, IQ (all rs < .33, all p > .09), education (all rs < .24, all p > .22), and chlorpromazine (all |rs| < .17, all p > .44) did not correlate with performance on any tasks.
a
Perceived Motion Direction Change
30 r = 0.67 p < .001 20
b
Perceived Orientation Change
6 r = 0.49 p = .01
10
2 Orientation 0 5 10 15 20 25
0 0 5 10 15
Motion 20 25
c
Perceived Orientation Change
6 r = 0.46 p = .02
BPRS Score
d
Perceived Orientation Change
6 r = -0.48 p = .01
BPRS Score
2 Orientation 0 10 20 30
Illness Duration
Fig. 4.The relationship between clinical measures in SZ and magnitudes of orientation and motion repulsion effects. Correlations between Brief Psychiatric Rating Scale (BPRS) symptoms ratings and (a) perceived motion direction changes in the motion repulsion task and (b) perceived orientation changes in the orientation repulsion task. In both cases, individuals with schizophrenia (SZ) with higher BPRS scores were more likely to have stronger repulsion effects. Correlations remained significant when excluding one potential BPRS outlier (motion: r = .67, p < .001; orientation: r = .49, p = .01). Box plots illustrate the range of motion (left) and orientation (right) repulsion in controls (CO; middle horizontal line denotes the median; the box boundaries represent the quartiles; and the whiskers range from the minimum to maximum values). Correlations were also calculated between magnitude of orientation repulsion and (c) illness duration and (d) social functioning score. SZ with stronger orientation repulsion were more likely to have poorer social functioning and longer duration of illness. Overall, poorer clinical measures predicted stronger perceptual repulsion effects.
12
Yang et al. reference presentation. Previous studies (Barch et al., 2012; Dakin et al., 2005) used a two-interval procedure with relatively short stimulus durations (500 ms), which may be more susceptible to attention deficits. Our results extend other reports that demonstrate various impairments in contrast processing in SZ (Kantrowitz, Butler, Schecter, Silipo, & Javitt, 2009; Must, Janka, Benedek, & Kri, 2004; ODonnell et al., 2002; but see Chen et al., 2003). For example, SZ fail to show typical contrast sensitivity modulations in the presence of collinear flanking stimuli (Must et al., 2004). Surround contrast modulation is believed to reflect gain control mechanisms in the primary visual cortex (Chubb et al., 1989; Lotto & Purves, 2001). Our results of relatively normal contextual modulation of luminance and impaired modulation of contrast together may suggest that abnormal visual processing in SZ begins at the level of the primary visual cortex. Furthermore, dysfunctional inhibitory circuitry in the early visual cortex may be central to deficits in contextual contrast modulation. Yoon et al. (2010) reported that SZ exhibited lower concentrations of GABA in the visual cortex that were also predictive of weakened contrast gain control, as evidenced by reduced orientation-specific surround suppression. Furthermore, these results could not be explained by a potential antipsychotic medication effect. We found that the chlorpromazine equivalent dose, which indexes dopamine D2 binding, did not correlate with performance on the surround contrast task (r = .12, p = .58) or any other contextual task (all |r| < .17, all p > .44). Moreover, atypical antipsychotic medications reportedly have little influence on visual sensitivity (Chen et al., 2003). Although it seems unlikely that medication contributed significantly to the weakened contrast illusion in SZ, further research is needed to eliminate this possibility completely.
Discussion
In this study, we examined contextual interactions in SZ across a broad range of visual tasks. SZ showed a weak contextual effect of contrast and, consequently, more accurate task performance relative to age- and gender-matched CO. Yet, the magnitude of contextual modulations associated with luminance, size, orientation, and motion were similar between groups. However, the strength of orientation repulsion and motion repulsion was related to clinical symptom severity and social functioning: Stronger repulsion effects were associated with greater psychiatric symptoms, longer duration of illness, and poorer social outcome. Moreover, with regard to overall contextual modulation strength, the two groups could not be distinguished. In summary, our findings argue against a general contextual processing deficit in SZ and instead point to a more precise problem; that is, abnormal contextual interactions in SZ are observed in specific visual submodalities and may be modulated by illness severity.
13 knowledge, this is the first investigation of orientation repulsion in SZ. Recently, Gold et al. (2012) examined the clinical correlates of three cognitive measures in a large cohort of SZ, including contextual contrast modulation. These measures were chosen for their potential use in clinical studies of SZ. The researchers found that the surround contrast illusion failed to predict clinical symptoms or social functioning, which is in agreement with our results. However, we found that the magnitudes of both motion and orientation repulsion illusions strongly predicted the severity of a range of clinical symptoms and the orientation illusion further predicted social functioning. Further investigation will be necessary to ascertain the usefulness of these particular contextual illusions for clinical studies of SZ.
14 underscore the need for more research. Finally, we report for the first time the profile of contextual processing in schizophrenia wherein we not only confirm older results but report new abnormalities as well. In sum, our study shows clear evidence arguing against the notion of a generalized deficit in contextual processing and further demonstrates the specificity and complexity of contextual deficits in schizophrenia.
Yang et al.
Birchwood, M., Smith, J., Cochrane, R., Wetton, S., & Copestake, S. (1990). The Social Functioning Scale: The development and validation of a new scale of social adjustment for use in family intervention programmes with schizophrenic patients. British Journal of Psychiatry, 157, 853859. Braddick, O. (1993). Segmentation versus integration in visual motion processing. Trends in Neurosciences, 16, 263268. Brainard, D. H. (1997). The Psychophysics Toolbox. Spatial Vision, 10, 433436. Butler, P. D., & Javitt, D. C. (2005). Early-stage visual processing deficits in schizophrenia. Current Opinion in Psychiatry, 18, 151157. Butler, P. D., Silverstein, S. M., & Dakin, S. C. (2008). Visual perception and its impairment in schizophrenia. Biological Psychiatry, 64, 4047. Chen, Y. (2011). Abnormal visual motion processing in schizophrenia: A review of research progress. Schizophrenia Bulletin, 37, 709715. Chen, Y., Levy, D. L., Sheremata, S., Nakayama, K., Matthysse, S., & Holzman, P. S. (2003). Effects of typical, atypical, and no antipsychotic drugs on visual contrast detection in schizophrenia. American Journal of Psychiatry, 160, 17951801. Chen, Y., Norton, D., & Ongur, D. (2008). Altered center-surround motion inhibition in schizophrenia. Biological Psychiatry, 64, 7477. Chubb, C., Sperling, G., & Solomon, J. A. (1989). Texture interactions determine perceived contrast. Proceedings of the National Academy of Sciences, USA, 86, 96319635. Churan, J., Khawaja, F. A., Tsui, J. M., & Pack, C. C. (2008). Brief motion stimuli preferentially activate surround-suppressed neurons in macaque visual area MT. Current Biology, 18, R1051 R1052. Cohen, J. P., & Servan-Schreiber, D. (1992). Context, cortex and dopamine: A connectionist approach to behaviour and biology in schizophrenia. Psychological Review, 99, 4577. Dakin, S., Carlin, P., & Hemsley, D. (2005). Weak suppression of visual context in chronic schizophrenia. Current Biology, 15, R822R824. Delord, S., Ducato, M. G., Pins, D., Devinck, F., Thomas, P., Boucart, M., & Knoblauch, K. (2006). Psychophysical assessment of magno- and parvocellular function in schizophrenia. Visual Neuroscience, 23, 645650. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2002). Structured clinical interview for DSM-IV-TR Axis I disorders, research version. New York: Biometrics Research. Gold, J. M., Barch, D. M., Carter, C. S., Dakin, S., Luck, S. J., MacDonald, A. W., III, . . . Strauss, M. (2012). Clinical, functional, and intertask correlations of measures developed by the Cognitive Neuroscience Test Reliability and Clinical Applications for Schizophrenia Consortium. Schizophrenia Bulletin, 38, 144152. Hamburger, K., Hansen, T., & Gegenfurtner, K. R. (2007). Geometricoptical illusions at isoluminance. Vision Research, 47, 32763285. Kantrowitz, J. T., Butler, P. D., Schecter, I., Silipo, G., & Javitt, D. C. (2009). Seeing the world dimly: The impact of early visual
Conclusion
This study presents a comprehensive evaluation of contextual interactions in SZ based on a battery of well-established psychophysical tasks that tap into distinct visual mechanisms. Our findings provide strong, new evidence against a broad, general, and unitary contextual processing deficit in schizophrenia. Given the current state of the literature, our findings represent a significant and important progress toward a more nuanced understanding of perceptual abnormalities in schizophrenia. Accurate perception forms the basis for our interactions with the external environment. Anomalous perceptual experiences, therefore, could lie at the core of the chronic difficulties confronted by individuals with schizophrenia as they attempt to navigate the complex, fast-moving, and constantly changing world. Authors Note
Eunice Yang and Duje Tadin are equally contributing authors.
Funding
This work was supported by grants from the National Alliance for Research on Schizophrenia and Depression, the National Institutes of Health (MH073028, EY007135, EY001319, EY007125, and EY019295), the National Institute of Child Health and Human Development (P30 HD15052) to the Vanderbilt Kennedy Center for Research on Human Development, and the World Class University Program through the Korea Science and Engineering Foundation, funded by the Ministry of Education, Science and Technology (R31-10089).
References
Albright, T. D., & Stoner, G. R. (2002). Contextual influences on visual processing. Annual Review of Neuroscience, 25, 339379. Andreasen, N. C. (1983). The Scale for the Assessment of Negative Symptoms (SANS). Iowa City: University of Iowa. Andreasen, N. C. (1984). The Scale for the Assessment of Positive Symptoms (SAPS). Iowa City: University of Iowa. Barch, D. M., Carter, C. S., Dakin, S. C., Gold, J., Luck, S. J., MacDonald, A. W., III, . . . Strauss, M. E. (2012). The clinical translation of a measure of gain control: The contrast-contrast effect task. Schizophrenia Bulletin, 38, 135143.
15
Pelli, D. G. (1997). The VideoToolbox software for visual psychophysics: Transforming numbers into movies. Spatial Vision, 10, 437442. Phillips, W. A., & Silverstein, S. M. (2003). Convergence of biological and psychological perspectives on cognitive co-ordination in schizophrenia. Behavioral and Brain Sciences, 26, 65138. Rossi, A. F., & Paradiso, M. A. (1999). Neural correlates of perceived brightness in the retina, lateral geniculate nucleus, and striate cortex. Journal of Neuroscience, 19, 61456156. Tadin, D., Kim, J., Doop, M. L., Gibson, C., Lappin, J. S., Blake, R., & Park, S. (2006). Weakened center-surround interactions in visual motion processing in schizophrenia. Journal of Neuroscience, 26, 1140311412. Tadin, D., Silvanto, J., Pascual-Leone, A., & Battelli, L. (2011). Improved motion perception and impaired spatial suppression following disruption of cortical area MT/V5. Journal of Neuroscience, 31, 12791283. Uhlhaas, P. J., & Mishara, A. L. (2007). Perceptual anomalies in schizophrenia: Integrating phenomenology and cognitive neuroscience. Schizophrenia Bulletin, 33, 142156. Uhlhaas, P. J., Phillips, W. A., Mitchell, G., & Silverstein, S. M. (2006). Perceptual grouping in disorganized schizophrenia. Psychiatry Research, 145, 105117. Uhlhaas, P. J., Silverstein, S. M., Phillips, W. A., & Lovell, P. G. (2004). Evidence for impaired visual context processing in schizotypy with thought disorder. Schizophrenia Research, 68, 249260. Yoon, J. H., Maddock, R. J., Rokem, A., Silver, M. A., Minzenberg, M. J., Ragland, J. D., & Carter C. S. (2010). GABA concentration is reduced in visual cortex in schizophrenia and correlates with orientation-specific surround suppression. Journal of Neuroscience, 30, 37773781. Yoon, J. H., Rokem, A. S., Silver, M. A., Minzenberg, M. J., Ursu, S., Ragland, J. D., & Carter, C. S. (2009). Diminished orientationspecific surround suppression of visual processing in schizophrenia. Schizophrenia Bulletin, 35, 10781084.